QUALITATIVE RESEARCH (Part 1)

QUALITATIVE RESEARCH



A central concept of the research designs discussed in previous chapters is that psychological constructs can be measured at the individual level and that understanding can be obtained by averaging these measurements over many persons. The participants respond to questions that in turn yield numbers that the researcher uses to form an understanding that is applicable to others. Participants may be subjected to various treatments or other manipulations, may complete inventories or instruments, and may interact with the researcher; all such actions, however, are used in the service of examining the relationship among the constructs assessed. Although we can learn much about counseling from the quantitative measurement of persons involved in the endeavor, as counselors well understand, the individual meaning that people attribute to their activities and experiences are critically important.
Qualitative research involves understanding the complexity of people's lives by examining individual perspectives in context. Qualitative research methodology is a radically different way to approach knowing and understanding. The following is part of a narrative given by Janie (pseudonym), a woman who was interviewed by the researcher, Lawless, at a women's shelter in Missouri. Janie described her childhood of living with her abused and depressed mother and her alcoholic father who moved their family away from civilization.
There was a time when I heard the gun go off in the basement of my home. . . . She didn't have a vehicle. She never had any money. He wouldn't let her have any money, and if she didn't have money, of course, she didn't have a way to go spend it, because she didn't have a vehicle, you know, she just didn't have any friends. [Crying] One time she just lost it. She went into a nervous breakdown. . . .
One day, and this just brings tears to my eyes to say, one day I remember I was a young teenager, about thirteen. My dad was working all the time, doing drugs, was an alcoholic, and my mom went through the same abuse I went through [with my husband], with my dad. She had a loaded .357 Magnum lying on the table and I walked in and she says, "See this gun?" I was scared to death of what she was going to do. She says, "I'm going to pick it up as soon as you walk out the door, and I'm going to shoot myself, unless you can give me ten reasons not to." And me being only thirteen, I had to give her ten reasons why. And that was a hard thing for me to do. (Lawless, 2001, p. 97)

Qualitative methodology emphasizes the importance of context in helping us understand a phenomenon of interest. Imagine how you would interpret Janie's coping mechanisms and relationship patterns as well as the impact of the past trauma on her life if you also knew the following facts from the inter¬view with her: Janie had to raise her own sister because later her father left and her mother was hospitalized for depression. Janie "impulsively got pregnant" at the age of 15. She talked about "how quickly her young husband became abusive," and how like her father, her husband "moved their mobile home ever farther out into the country, isolating her more and more." Despite all of the hardship that Janie encountered while growing up with a depressed mother and raising her sister on her own, she reported having "a very happy childhood overall" and a "very, very close" relationship with her mother. By the end of the interview, Janie described her mother as "a good woman," who "just did everything in the world to please anybody" (Lawless, 2001, pp. 96-97).
The researcher (Lawless) conducted face-to-face interviews with numerous female survivors of violence. She invites the readers to compare the narrative data generated through these interviews with data obtained from quantitative measures that include questions like: "1) Overall, how upset were you by this experience—extremely upset, somewhat upset, not very upset, or not at all upset?; 2) Looking back on it now, how much effect would you say this experience(s) has had on your life—a great effect, some effect, a little effect, or no effect?" (Russell, 1986, p. 138, as cited in Lawless, 2001).
The comparisons explicate the sharp differences in the type, range, and depth of the data generated through qualitative and quantitative methodologies. Janie's first-person account provided in-depth information about her lived experiences as a domestic violence survivor. Her narratives revealed the complex and inextricably intertwined relationships between life events, personal development, cognitive appraisals, psychological adjustment, and human resilience. Qualitative methodology stresses the process in which individuals create and give meanings to their social experience and lived realities. In contrast, quantitative methodology paints a broad picture of the relationship among the constructs assessed through generating and averaging nomothetic data over a relatively large number of participants.
This chapter will provide a discussion of qualitative research methods and designs. First, the definition of qualitative research and key myths and facts about qualitative research will be presented. Next, we will briefly review the philosophical foundations of qualitative research and the reasons why it is important to attend to these underlying paradigms. Furthermore, three strategies of qualitative inquiry (i.e., grounded theory, phenomenology, and consensual qualitative research) will be illustrated with exemplar studies. The primary tasks in qualitative research (ways to gather and analyze data) will be presented. Finally, the criteria that may be used for evaluating the rigor of qualitative methodology will be briefly described.

WHAT IS QUALITATIVE RESEARCH?
Qualitative research is employed in various disciplines (e.g., anthropology, education, nursing, sociology, and psychology) with numerous strategies of inquiry (e.g., ethnography, grounded theory, discourse analysis). Students often feel confused about and overwhelmed by the variation of terminology as well as the philosophical and procedural diversity of qualitative research. For this reason, we will delineate several common myths and basic facts about the nature of qualitative research in the following sections.

DEFINING QUALITATIVE RESEARCH
A generic definition of qualitative research is provided by Denzin and Lincoln (20001 as flows:
Qualitative research is a situated activity that locates the observer in the world. It consists of a set of interpretive, material practices that make the world visible. These practices transform the world. They turn the world into a series of representations, including field notes, interviews, conversations, photographs, recordings, and memos to the self. At this level, qualitative research involves an interpretive, naturalistic approach to the world. This means that qualitative researchers study things in their natural settings, attempting to make sense of, or to interpret, phenomena in terms of the meanings people bring to them.
Qualitative research involves the studied use and collection of a variety of empirical materials—case study; personal experience; introspection; life story; interview; artifacts; cultural texts and productions; observational, historical, interactional, and visual texts—that describe routine and meanings in individuals' lives. Accordingly, qualitative researchers deploy a wide range of interconnected interpretive methods, hoping always to get a better understanding of the subject matter at hand. It is understood, however, that each practice makes the world visible in a different way. Hence, there is frequently a commitment to using more than one interpretive practice in any study. (pp. 3-4)

In other words, qualitative researchers are devoted to understanding the specifics of particular cases and embedding their research findings in an ever-changing world. Influenced by the interpretivist-constructivist tradition, qualitative researchers believe that objective reality can never be fully understood or discovered, and there are many possible ways of looking at realities. Qualitative researchers are interested in capturing the individual's point of view through the use of multiple strategies such as interviews and observations (emit and idiographic perspectives), instead of deploying etic and nomothetic approaches that emphasize the goal of discovering and describing universal principles by quantifying the observed phenomena. Qualitative researchers choose from a variety of research tools in accordance with their research questions and contexts to better understand the phenomenon of interest (Nelson, Treichler, & Grossberg, 1992), rather than verifying or falsifying an a priori hypothesis through experimental designs and statistical analysis. They value rich descriptions of the phenomenon under analysis and attempt to represent an individual's lived experience through writing and interpretations.
Similarly to quantitative research, the questions that qualitative researchers ask their participants and the methods that they utilize to observe certain phenomena are all "filtered" through the researchers' lenses of knowledge, language, values, and worldviews. Denzin and Lincoln (1998) described qualitative research as "an interactive process" shaped by the researcher's "personal history, biography, gender, social class, race and ethnicity, and those of the people in the settings" (p. 4). Qualitative researchers acknowledge ("bracket") their assumptions about the study by taking field notes, writing reflexive journals, and informing the readers as to what their "filters" are. The lived experiences of the research participants are what qualitative researchers focus on, and the researchers are the instruments in this discovery process who hold their expectations and hunches about the phenomenon under study in abeyance (Rennie, 2000).
The discovery orientation of the qualitative approach helps to focus on the context where the phenomenon is situated and make the findings more applicable for people's everyday lives in various cultures. Qualitative inquiry allows researchers to study the local interactions in counseling settings and their meanings for counselors and clients. For these reasons, qualitative inquiry is particularly appropriate for multicultural/cross-cultural research (see Morrow, Rakhasha, & Castaneda, 2001) as well as process and outcome research (see McLeod, 2001). In fact, it has gained increasing popularity among researchers in different parts of the world, such as Asia (e.g., Kim & Cho, 2005) and Europe (e.g., Forum Qualitative Sozialforschung).

MYTHS AND FACTS ABOUT QUALITATIVE RESEARCH
Due to insufficient understanding of the definition and characteristics of qualitative research, there are several common myths about this form of inquiry. First, many students equate "qualitative data" with "qualitative research," thereby holding a mistaken belief that qualitative research consists only of asking people open-ended questions and analyzing participants' answers. Second, some people think that there is only one kind of qualitative methodology and are unaware that there are a variety of methods in qualitative research, such as phenomenology and narratology. Third, some people believe that qualitative research should be used only in areas where we do not have enough information to do quantitative studies; this perspective overlooks how qualitative research can add to the depth and breadth of our understanding about certain phenomena. Finally, some students choose to adopt a qualitative approach because they feel uncomfortable with statistics or believe that it is easier to conduct a qualitative inquiry than a quantitative one. These students ignore the fact that conducting qualitative research actually requires rigorous efforts and introspection; the researcher listens to other people's stories and retells the stories in a way that she or he understands them or even reconstructs the story with the participants.
Creswell (1998) indicated that qualitative researchers should be willing to: (1) "commit to extensive time in the field"; (2) "engage in the complex, time-consuming process of data analysis the ambitious task of sorting through large amounts of data and reducing them to a few themes or categories"; (3) "write long passages, because the evidence must substantiate claims and the writer needs to show multiple perspectives"; and (4) "participate in a form of social and human science research that does not have firm guidelines or specific procedures and is evolving and changing constantly" (pp. 16-17). Although qualitative procedures may not be as clearly delineated as quantitative procedures, researchers must acquire highly specialized knowledge and demonstrate rigorous endeavors, as will be made evident in the remaining parts of this chapter.
Students are encouraged to reflect on the following factors before deciding to conduct a qualitative study: (1) the fit between the research question and qualitative methodology; (2) the extent of their knowledge on the fundamental paradigms and methods of qualitative inquiry and the level of skills gained from appropriate coursework and research apprenticeship; (3) whether they have adequate support from advisors and/or research mentors who are knowledgeable about qualitative methodology; (4) the existing knowledge bases and types of research designs previously used; and (5) their readiness to conduct a rigorous, qualitative investigation. These elements are worthy of considerable reflection. In the subsequent sections of this chapter, we will explore some of the essential components of qualitative research.

PHILOSOPHICAL FOUNDATIONS
In Chapter 1, we contrasted four paradigms that bear on the research process. Readers are encouraged to review that chapter again to be familiar with the philosophical foundations of scientific inquiry.
Overall, quantitative research is aligned with two of the paradigms-constructivism and critical theory. In general, all qualitative paradigms assume relativist ontology (there are multiple realities that are socially and individually constructed) and transactional epistemology (the knower and the known are inextricably intertwined), as well as dialogic/interpretive methodology (Guba & Lincoln, 1998). It is crucial to understand these underlying paradigms so that they match the researcher's own personal values, beliefs, and personality, as well as personal and mental models (Morrow et al., 2001). Also, the research approach adopted should be appropriate for answering research questions within the context of existing knowledge. As introduced later in this chapter, there are many different strategies of inquiry in qualitative research; each inquiry has somewhat different philosophical underpinnings (see Ponterotto, 2005b, for an excellent discussion about locating the strategies of inquiry within research paradigms). Therefore, understanding the basic tenets and philosophical foundations of qualitative research will help students select a particular paradigm and strategy of inquiry that may best address a particular area of inquiry.
Readers who are interested in advancing their knowledge about the philosophical underpinnings of these paradigms can find excellent introductions and discussions in the Handbook of Qualitative Research (e.g., Lincoln & Guba, 2000; Schwandt, 2000), the Handbook of Counseling Psychology (Morrow & Smith, 2000), the Handbook of Multicultural Counseling (Morrow et al., 2001), and the special issue on qualitative research in the Journal of Counseling Psychology (e.g., Ponterotto, 200513). Next, we will shift our focus to qualitative research methods and design issues.

STRATEGIES OF INQUIRY
Each qualitative strategy of inquiry is connected to specific paradigms and research designs. Researchers should be clear about the differences among various strategies of inquiry in order to make informed decisions regarding what qualitative approaches to use and when to use them (Creswell, 1998), and subsequently to design their studies according to the guidelines of a particular chosen strategy. The rationale of using a specific paradigm and strategy of inquiry should be presented and the relationship between the paradigms/strategies of inquiry and purposes/research questions of the studies should be explicated in the final write-up.
It is worth mentioning that great confusion may arise at this stage because a variety of strategies of inquiry exist across various disciplines. For example, Strauss and Corbin (1990), who are in the field of sociology and nursing, grouped qualitative strategies of inquiry into five categories (grounded theory, phenomenology, life history, ethnography, and conversational analysis); conversely, Moustakas (1994), who is in the field of psychology, listed six qualitative traditions (grounded theory, hermeneutics, empirical phenomenological research, ethnography, heuristic research, and transcendental phenomenology). Narrative studies provides another example of this diversity: According to Hoshmand (2005), "narratology is a term historically used to refer to the study of narratives in the literary field, though other disciplines in the humanities (such as history) and the social sciences (such as cultural studies) also are associated with the study of narratives"(p. 178). Hoshmand further clarified that she used the term narratology "as a way of distinguishing a mode of qualitative inquiry and data analysis that is informed by narrative theory," differing from "other qualitative research that involves narrative data but not a narrative perspective per se" (p. 178).
Each of these numerous types of qualitative research has distinct purposes and methods. Figure 11.1, borrowed from Tesch (1990), provides a schematic diagram for various types of research depending on the research interest. The primary categories in this hierarchy, which are based on the interest of the research, focus on (1) the characteristics of language, (2) the discovery of regularities, (3) the comprehension of the meaning of text or action, or (4) reflection.
Due to space limitations, we selected three strategies of inquiry to present in this chapter based on their potential utility in counseling research: grounded theory, phenomenology, and consensual qualitative research. We will briefly describe these three strategies of inquiry along with exemplar studies. Note that the studies cited in this chapter are good examples of qualitative research but are not the "only correct" way to conduct qualitative research. These three types of strategies are mentioned to provide a flavor of the variations possible; readers interested in a particular approach will need to explore the specific methodological literature for that type. Also, students who use these strategies of inquiry for theses or dissertations may also examine the illustrations and examples of writing different sections of qualitative studies in Wang, Heppner, and Heppner (2004a, 20046).

GROUNDED THEORY
This section first provides the definition and purpose of the grounded theory approach. Then, its primary features—(1) constant comparative method, (2) memo writing, (3) theoretical sampling, and (4) the emerging theory that is grounded in data—will be discussed and elucidated with exemplar studies.
The grounded theory approach was developed by Glaser and Strauss (1967) in an attempt to challenge the dominant trend of "excessive reliance on the quantitative testing of hypotheses derived from a small number of grant (totalizing) theories, typically through numerical survey and other statistical approaches" (Henwood Pidgeon, 2003, p. 132). According to Charmaz (2000),
Grounded theory methods consist of systematic inductive guidelines for collecting and analyzing data to build middle-range theoretical frameworks that explain the collected data. Throughout the research process, grounded theorists develop analytic interpretations of their data to focus further data collection, which they use in turn to inform and refine their developing theoretical analyses. (p. 509)

The grounded theory approach is rooted in sociology and the tradition of symbolic interactionism. It is appropriate for studying "the local interactions and meanings as related to the social context in which they actually occur," and therefore is particularly attractive to psychologists (Pidgeon, 1996, p. 75). In fact, the grounded theory methods have gained increasing popularity among researchers across various disciplines over the past two decades (see Fassinger, 2005; Rennie, Watson, & Monteiro, 2002) and has been named "the most influential paradigm for qualitative research in the social sciences today" (benzin, 1997, as cited in Patton, 2002, p. 487).
Grounded theory has been associated with seemingly opposing philosophical views—"realism (by claiming to directly reflect the `data')" and "constructivism (inherent to the approach of symbolic interactionism)" (Henwood & Pidgeon, 2003, p. 134). Related to this epistemological tension, grounded theorists have been advocating competing approaches to data analysis and interpretations. For example, Strauss and Corbin (1990, 1998) promote the use of open coding, axial coding, and selective coding in specifying the properties and dimensions of categories and organizing the emerging theory in a conditional matrix that contains the antecedents, context, intervening conditions, and consequences of core categories. Glaser (1992) stresses the importance of constant comparative method and theoretical memoing in generating a theoretical model. Rennie (2000) presents the grounded theory method as methodical hermeneutics (i.e., a methodological approach that "involves the interplay between induction and abduction conducted reflexively") (p. 494). Recently, the constructivist version of grounded theory also has been advocated by several grounded theorists (e.g., Charmaz, 2000; Henwood & Pidgeon, 2003), which emphasizes the interpretive process of a grounded theory study.
Novice researchers often are perplexed by these divergent perspectives. Readers should consult the key literature listed earlier for more information about specific grounded theory methods. Nonetheless, there are some commonalities across these various grounded theory methods, which will be described in the following sections.

CONSTANT COMPARATIVE METHOD
The constant comparative method, the hallmark of grounded theory research, consists of four stages: (1) comparing incidents applicable to each category, (2) integrating categories and their properties, (3) delimiting the theory, and (4) writing the theory (Glaser & Strauss, 1967). It is worth noting that although one stage leads to another, some earlier stages will continue operating simultaneously until the termination of the data analysis.
Glaser and Strauss (1967) developed the constant comparative method to generate "many categories, properties, and hypotheses about general problems" and to formulate a theory that is grounded in the data, instead of "[ascertaining] either the universality or the proof of suggested causes or other properties" (p. 104). Specifically, the following procedure is suggested for generating categories and properties: (1) assign codes to segments of the text (meaning units) in the initial open coding phase; (2) sort these codes into clusters according to their shared meanings; (3) convert the meaning of each cluster into a category; and (4) discontinue the sorting of the codes when no new category emerges (i.e., categorization has reached saturation) (Rennie, 2000).
Pidgeon and Henwood (1996) emphasized the importance of documenting this analytical process fully while conducting grounded theory research, which helps to track the procedures and helps the researchers become aware of their implicit, a priori assumptions. Pidgeon and Henwood described the analytic process using the flow chart shown in Figure 11.2.

Consistent with the depiction in Figure 11.2, this data collection and analytic procedure was described as a "zigzag" process by Creswell (1998), "out to the field to gather information, analyze the data, back to the field to gather more information, analyze the data, and so forth" (p. 57). In other words, data collection and data analysis are not discrete stages in grounded theory research. This method sets grounded theory apart from content/thematic analysis, which employs reliability and validity as criteria and uses "the counting of instances within a predefined set of mutually exclusive and jointly exhaustive categories" (see Pidgeon, 1996, p. 78).

MEMO WRITING
It is essential to keep memos during the process of conducting the constant comparative method. Memo writing is defined as a process to record "hunches; comments on new samples to be checked out; explanations of modifications to categories; emerging theoretical reflections; and links to the literature" (Pidgeon & Henwood, 1996, p. 95). Charmaz (2000) indicated,
Memo writing is the intermediate step between coding and the first draft of the completed analysis.... It can help us to define leads for collecting data----both for further initial coding and later theoretical sampling. Through memo writing, we elaborate processes, assumptions, and actions that are subsumed under our codes." (p. 517)

Pidgeon and Henwood (1996) also warned that researchers should "write a memo as soon as the thought has occurred, for, if left unrecorded, it is likely to be forgotten" (p. 95). In sum, memo writing is a vital technique for grounded theorists. It not only serves as an instrumental mechanism for the constant comparative practice, but also facilitates the theoretical sampling and theory development processes that will be described in the following sections.

THEORETICAL SAMPLING
The data sources of grounded theory research could include a combination of data types (e.g., archival/textual materials, participant observation, autobiographies, and journals). Among all of these options, however, interviews with the participants are a primary data source (Pidgeon & Henwood, 1996).
Theoretical sampling characterizes the ongoing analytic process in the field and is a theory-driven method of sampling. Theoretical sampling helps researchers target new data that would facilitate the emergence of theory after the initial analysis of the data at hand. It is used by grounded theorists “to select a sample of individuals to study based on their contribution to the development of the theory. Often, this process begins with a homogeneous sample of individuals who are similar, and, as the data collection proceeds and the categories emerge, the researcher turns to a heterogeneous sample to see under what conditions the categories hold true”. (Creswell, 1998, p. 243)
Therefore, theoretical sampling differs from the sampling method used in quantitative research, in which researchers are expected to obtain a representative sample in order to enhance generalizability of the research finding. "Theoretical saturation" occurs when "the new data fit into the categories already devised" (Charmaz, 2000, p. S20) and it indicates the ending of the data collection process.

THEORY IS GROUNDED IN THE DATA
Grounded theorists argued that theories should be derived from the data. Unlike quantitative researchers, grounded theorists neither test an existing theory nor try to fit their data into preconceived concepts. Instead, all of the theoretical concepts should be derived from the data analysis and account for the variation in the studied phenomenon, thereby allowing the theoretical framework to emerge through the aforementioned constant comparative practice, memo writing, and theoretical sampling processes (Charmaz, 2000).

EXEMPLAR STUDY
We will use a grounded theory study conducted by Morrow and Smith (1995) as an example to illustrate the grounded theory method. This study aims at understanding the survival experience of women survivors of childhood sexual abuse and representing their coping processes in a theoretical framework that emerged from the data. The authors explicitly dis¬cussed the reasons for choosing a qualitative methodology and grounded theory approach for the purpose of their study.
As Hoshmand (1989) noted, qualitative research strategies are particularly appropriate to address meanings and perspectives of participants. In addition, she suggested that naturalistic methods offer the researcher access to deep-structural processes. . . . The primary method of investigating those realities was grounded theory (Glaser & Strauss, 1967), a qualitative research method designed to aid in the systematic collection and analysis of data and the construction of a theoretical model. . . . Chosen to clarify participants' understandings of their abuse experiences, the methods used involved (a) developing codes, categories, and themes inductively rather than imposing predetermined classifications on the data (Glaser, 1978), (b) generating working hypotheses or assertions (Erickson, 1986) from the data, and (c) analyzing narratives of participants' experiences of abuse, survival, and coping. (pp. 24-25)

Participants for this study included 11 childhood sexual abuse survivors with a wide age range (25 to 72 years old) and diverse racial/ethnic backgrounds, sexual orientations, and educational levels; three of the participants also reported physical disabilities. Morrow and Smith (1995) provided a detailed description of the participants' backgrounds (including brief abuse histories and counseling experiences). These "thick descriptions" (a qualitative tactic that will be further explained later in this chapter) provide the readers an understanding of the context of the phenomenon. The authors also did a nice job of describing the participant recruitment process: Letters that detailed the research's purpose and process were sent to therapists with expertise in working with sexual abuse survivors; these therapists were asked to give the research announcement to those clients "who might benefit from or be interested in participating in the study" (p. 25).
Morrow and Smith (1995) clearly identified their sampling criterion: participants' "self-identification as an abuse survivor" based on the constructivist approach (i.e., "[accepting] the stories of participants at face value as their phenomenological realities") (p. 25). The authors utilized a variety of data sources, including: (1) semi-structured, 60- to 90-minute, in-depth interviews with individual participants; (2) a 10-week focus group with 7 of the 11 research participants that emphasized the survivors' coping experiences and thoughts on the emerging categories from the initial data analyses; (3) documentary evidence such as participants' journals and artistic productions; and (4) Morrow's self-reflective and analytic memos, which consisted of the researcher's reflections and assumptions about the data. The interviews were transcribed verbatim, and "the data corpus consisted of over 2,000 pages of transcriptions, field notes, and documents shared by participants" (p. 25).
Following Strauss and Corbin's (1990) suggestions, the authors conducted open coding, axial coding, and selective coding on the data, and "the language of the participants guided the development of code and category labels, which were identified with short descriptors, known as in vivo codes, for survival and coping strategies" (Morrow & Smith, 1995, p. 26). Subsequently, the authors generated a theoretical framework for the surviving and coping process experienced by those childhood sexual abuse survivors. Figure 11.3 shows the theoretical model derived from the data analysis.
The authors provided a rich description of the model, which serves as an overall illustration of the phenomenon under study. They also elucidated the categories in the model with quotes from the participants. For example, the following is the description of a "Phenomena" category:
[Participants experienced what was termed helplessness, powerlessness, and lack of control. Lauren provided an exemplar of the second category, illustrating the pervasiveness of her perpetrator's power:
He stands there. A silhouette at first and then his face and body come into view. He is small, but the backlighting intensifies his figure and he seems huge, like a prison guard. He is not always there but it feels like he might as well be. When he's not there, I search the distance for him and he appears. He seems to be standing there for hours. As if he's saying, you are weak, I am in control.
Not only did Lauren experience powerlessness during her abuse, but her lack of control invaded her dreams and her moments alone. (pp. 27-28)
Furthermore, the authors demonstrated that the analytic process is not only sequential, but also recursive, which is the spirit of the constant comparative method and theoretical sampling. They described how the researcher's memos facilitated this process:
Meghan foreshadowed one of these phenomena the first night of the group, when she said, "To keep from feeling my feelings, I have become a very skilled helper of other people." Throughout the data, others echoed her words. The analytic moment in which this category emerged is illustrated in the following analytic memo written by Morrow (in vivo codes are in italics):
I'm reading a higher level of abstraction. Is the overarching category protection from feelings? Many categories are subsumed under it: One talks to get out the stories; the feelings are less intense. Fake orgasm (sex) because you don't have any physical feelings. Art was used to deal with feelings, express anger, release the pressure of the feelings, use chemicals to deal with feelings (and a whole complex interaction here) .. .
Existing and emergent codes and categories were compared and contrasted with this category; the category was modified to accommodate the data, producing the phenomenon that was labeled being overwhelmed by threatening or dangerous feelings—feelings that participants described as subjectively threatening or dangerous. (p. 27)

Morrow and Smith (1995) also explicitly explained the rigor of their method: triangulating multiple data sources (i.e., interviews, focus groups, documents, memos), immersing in the data (the entire data collection period lasted over 16 months), member check (participants were invited to verify data analysis results and revise the emergent theoretical model), and peer debriefing (Morrow invited other qualitative researchers to review the analysis, her writing, and the "audit trail that outlined the research process and evolution of codes, categories and theory" [p. 261).
Other exemplar grounded theory studies include Rennie and his colleagues' research on psychotherapy (Rennie, 1994a, 1994b, 1994c; Watson & Rennie, 1994) and Fassinger and her colleagues' studies on career development of highly achieving Black and White American women (Richie et al., 1997) and highly achieving women with disabilities (Noonan et al., 2004). Interested readers may also find the researchers' own thoughts about their investigation process in their articles on the grounded theory methods (e.g., Fassinger, 2005; Rennie, 1996).


PHENOMENOLOGY

The purpose of phenomenology is "to produce an exhaustive description of the phenomena of everyday experience, thus arriving at an understanding of the essential structures of the 'thing itself', the phenomenon" (McLeod, 2001, p. 38). In the following paragraphs, the historical development of the phenomenological approach and its major principles will be introduced along with exemplar studies.
Phenomenology has its roots in the work of Edmund Husserl on phenomenological philosophy. Since then, many individuals (e.g., Giorgi, Moustakas, Polkinghorne) in the social science field have followed its tenets and transferred Husserl's work from philosophy to psychological or sociological research. Over the past decades, different approaches to phenomenology were developed (e.g., empirical/psychological phenomenology, hermeneutic phenomenology, existential phenomenology, social phenomenology, reflective/ transcendental phenomenology, and dialogical phenomenology). (For a review, see Creswell, 1998.) McLeod (2001) provided an overview of the three branches of "new" phenomenological traditions that evolved from Husserlian phenomenological principles: (1) "the Duquesne school of empirical phenomenology" developed by Giorgi and his colleagues in North America; (2) "the method of 'conceptual encounter' developed by Rivera" who was influenced by the Duquesne school and "the German social psychologist Lewin"; and (3) "the existential-phenomenological investigations" of "the Scottish psychiatrist Laing and his colleagues" at the Tavistock Institute for Human Relations in England (pp. 40-48). Interested readers are referred to McLeod for a comparison between the original phenomenological philosophy developed by Husserl and the "new" phenomenological traditions advocated by psychological researchers. Also, a recent review of the phenomenological movement in psychology and a comparison between phenomenological research and other qualitative methods can be found in Wertz (2005).
In spite of the existence of multiple phenomenological approaches, phenomenological researchers follow some general guidelines in developing plans for their studies. These guidelines were summarized by Creswell (1998) into five areas: (1) philosophical perspectives and epoche, (2) research question and lived experiences, (3) criterion-based sampling, (4) phenomenological data analysis, and (5) essential, invariant structure (or essence) of the lived experience. These five dimensions of the guidelines are key concepts of the phenomenological approach and will be illustrated below with an exemplar study.

PHILOSOPHICAL PERSPECTIVES AND EPOCHE
First and foremost, researchers who intend to adopt the phenomenological approach should have a thorough understanding of the underlying philosophy because its philosophical assumptions greatly influence the formulation of research questions and inquiry techniques (Creswell, 1998; McLeod, 2001). Because there are various phenomenological traditions, as mentioned earlier, the philosophical perspectives embraced by the researchers should be specified in the final write-up. Furthermore, before conducting a phenomenological study, researchers should "bracket" (i.e., set aside) their assumptions and judgments about the phenomenon, and these presuppositions should be documented. This strategy was termed epoche by Husserl. Phenomenological researchers use epoche to bracket and question all of their assumptions, hoping to reveal new and meaningful understanding that transcends the extant knowledge about a particular phenomenon (McLeod, 2001).

RESEARCH QUESTIONS AND LIVED EXPERIENCES
The key subject matter studied by phenomenological researchers is "the lived world" of human beings (i.e., "the life-world manifests itself as a structural whole that is socially shared and yet apprehended by individuals through their own perspectives") (Wertz, 2005, p. 169). 4 phenomenological study aims to "fill in the gap" between "knowledge and reality that requires qualitative knowledge, that is, an under¬standing of what occurs" (Wertz, 2005, p. 170). The research questions for a phenomenological study are developed in order to understand the everyday lived experiences of individuals and to explore what those experiences mean to them (Creswell, 1998). Data are typically generated through written or verbal responses of the research participants to "a descriptive task with instructions" or an open-ended interview (Wertz, 2005, p. 171). The research questions guide the development of the descriptive task or specific interview questions. Long interviews are often utilized to generate in-depth dialogues between the researcher and the participant about a specific phenomenon. Wertz suggested that the best type of data for phenomenological researchers is "concretely described psychology life," which helps to further our understanding about a given phenomenon beyond "any previous knowledge or preconceptions" (p. 171).

CRITERION-BASED SAMPLING
Criterion-based sampling is often used in a phenomenological study to select participants who meet the following criteria: (1) they experienced the phenomenon under study, and (2) they can articulate their lived experiences (Creswell, 1998). Similar to the grounded theory approach, there is no absolute number of participants needed for a phenomenological study. Yet, in assessing the adequacy of the sample size, the following criteria were proposed by Wertz (2005): "deliberation and critical reflection considering the research problem, the life-world position of the participant(s), the quality of the data, and the value of emergent findings with regard to research goals" (p. 171).

PHENOMENOLOGICAL ANALYSIS
Husserl developed the procedure of intentional analysis, which "begins with a situation just as it has been experienced—with all its various meanings—and reflectively explicated the experiential processes through which the situation is lived," thereby producing "knowledge of human situations, their meaning, and the processes that generate those meanings" (Wertz, 2005, p. 169). The intentional analysis is carried out through the use of epoche and an empathic understanding of the participants' lived experiences. In the analytic process, the researcher constantly "focuses on relations between different parts of the situation and the psychological processes that subtend it while attempting to gain explicit knowledge of how each constituent contributes to the organization of the structure as a whole" (Wertz, 2005, p. 172). Yet, phenomenological researchers who follow the different traditions mentioned earlier have proposed different analytic techniques (McLeod, 2001). Researchers need to decide which tradition to follow when conducting their analyses and clearly state the steps of the analytic method in their research methodology section.

ESSENCE OF THE EXPERIENCES
Husserl advocated the use of the intuition of essence or eidetic reduction to understand the essence of the phenomenon under study and to achieve psychological reduction. For this purpose, he developed a procedure, free imaginative variation, which "starts with a concrete example of the phenomenon of which one wishes to grasp the essence and imaginatively varies it in every possible way in order to distinguish essential features from those that are accidental or incidental" (Wertz, 2005, p. 168). The exhaustive descriptions of the phenomenon of interest are subsequently constructed through extracting the significant statements, formulating meanings, and clustering themes of the original data (Creswell, 1998). The exhaustive descriptions could provide the readers an understanding of the essence of the lived experiences and represent a unifying structure of a phenomenon.

EXEMPLAR STUDY
A study conducted by Teixeira and Gomes (2000) in Brazil will be used as an example of phenomenological studies. Their research aimed to understand the career trajectories of Brazilians who have chosen to change careers at least once in the past. The authors indicated that past research has not focused on the process of self-initiated career change, and most studies used quantitative methodology. The phenomenological approach was used for this study because it allowed the researchers to understand the meanings that participants ascribed to their autonomous career change in a specific context.
Teixeira and Gomes (2000) explicitly described the criterion they used to select participants: All of the participants would have experienced at least one self-initiated career change. Specifically, the participants were chosen "who had left one career for which they had received a university training for another that also required advanced training, such that the change had been voluntary (criteria for time spent in the careers were not established)" (p. 81). A "thick" (detailed) description of the participants (e.g., their first and current careers) was provided in the article.
The authors utilized interviews as the data collection method for this study. They started the interviews with a broad question, "I should like you to tell me about your career choices," and left it open-ended for the participants to freely discuss their career trajectories. Interviews were conducted with seven participants (four women and three men; age range from 32 to 42 years), audio taped, and subsequently transcribed. Locations (i.e., participants' home or work settings) and the actual lengths (i.e., one to two and a half hours) of the interviews were described in their manuscript.
Teixeira and Gomes (2000) followed the Duquesne school of empirical phenomenology tradition and explicitly delineated the phenomenological analytic approach utilized in their study:
The analytical procedures observed the methodological recommendations of Giorgi (1997) and Lanigan (1988) in the following way: transcriptions and successive readings of the material, search and demarcation of significant units within each interview, verification of cross units among interviews, and definition of the main themes. The relations between parts and whole were worked out by the preparation of a synthesis for each narrative. The confront [sic] between these syntheses and their constitutive themes were the base for the phenomenological description. Throughout this process, the researchers' prior knowledge of the subject was set to one side, thereby avoiding interferences from anticipated qualitative judgments (phenomenological epoche). The analysis rigorously observed the three steps of phenomenological method (description, reduction, interpretation) as indicated below. Excerpts taken from the interviews were used as illustrative evidence of the researcher' phenomenological understanding.

The authors concluded that their analysis yielded five essential themes that "comprise a general structure of the experiences described by the interviewees, although the limits, contents and scope of each theme may vary from case to case according to personal circumstances" (p. 82). Following the methodological recommendations of prominent phenomenological researchers, Teixeira and Gomes (2000) defined each theme (which they called a "moment") and subsequently identified the interviewees' accounts that "specify the affective, cognitive and conative limits of the experience that participants lived through in relation to each of the selected themes" (p. 82). The authors then provided a rich description of the themes with the participants' narratives. Following is the description of one of the five themes ("moments").
Moment S: Evaluating the change and present situation
The participants evaluated their professional careers, including their present situations, in a positive way. The evaluation was accompanied by a feeling of having matured and of personal transformation. Career change was generally associated with personal changes, for example the feeling of being less tied down and capable of making changes in other spheres of life (S2: ". . . after I took this decision I feel much more given to change, just as I had changed"; S4: ". . . I think that all that I had gone through made me succeed in not being so tied up with things .. . I am much more ... much lighter") or yet as a transformation in the personal out¬look, ceasing to like working to a "prepared formula" (S5).
Career change was also seen as a personal search for interpersonal development, self-determination and, most importantly, increased self-esteem (S7; ". .. it is something that makes me feel good, I like to hear people say that I am a doctor, I feel proud of what I am"). Yet others viewed their professional trajectory as an opportunity for steady development of their personality and of their capacities and interests, integrating all the professional experience they had (S3, S1).
Sl: "So I made it ... made use of knowledge, right, built up through a piece of my life in all the areas that I worked in, and I'm still reaping the benefit today. . . ."
The contrast with these positive evaluations on career changes, in which those interviewed considered themselves more mature and in the process of re-engagement in a profession, is given by S6. She still claims to have adolescent doubts about future career, just as she is on the point of finishing a course in Visual Arts (her professional interest abandoned in adolescence).
S6: "I think that basically I didn't change. I think I still . . . I'm 36 now, and I still feel I have those adolescent doubts.... 'What am I going to be in the future?' . . . I have the feeling that I'm still looking for something .. . [Can you see how you will be, I don't know, ten, fifteen years from now?] No... no, I can't, I can't, for me it's all far too hazy." (p. 87)

After summarizing and illustrating each theme that emerged from the data analyses, Teixeira and Gomes (2000) presented the essential structure (phenomenological reduction) of the phenomenon under study. They used the reduction to illustrate the characteristics of the phenomenon of career change. The complete description is approximately three pages long; below is its summary:
In summary, career change can be defined as a broad process of change in the relationship between the participants and their work. The change begins with the awareness of feelings of dissatisfaction with their career; leading the participants to ask questions of themselves and what they do. The questioning of career emerges in the consciousness as a figure whose background is the other roles lived by the participants in their daily lives. Personal plans are redefined and new priorities established, enabling the career to acquire a different meaning in this new context. Thus the search for a new career is also a search for a new way for the participants to express themselves in the world, which can lead them—although not necessarily—to feelings of greater self-determination and personal satisfaction. (p. 90)

In addition to the phenomenological reduction, Teixeira and Gomes also presented their interpretations of the phenomenon, which focus on "the possible meanings of these changes and of the career change itself" as follows (p. 91).
In general, the statements analyzed in this study suggest that career change can result from a career choice which we could call immature at adolescence. Immaturity is understood as a decision taken without the necessary exploration of alternatives which would allow the subjects to build up a clear occupational self-understanding and to make choices compatible with their interests and the conception that they have of themselves (Super, 1963; Super, Savickas, & Super, 1996).
But the histories presented by those interviewed also shows [sic] us that the choice made at adolescence is not something definitive. Above all it is a continuing process that reaffirms itself through practice of the profession or which provides experiences leading to new choices. Rather than being a fundamental error, as one of those interviewed called it, an inappropriate choice at adolescence can also be a way leading to deeper knowledge of oneself, and to personal and even professional changes.... This interpretation, moreover, is in contrast with the idea that adolescents (or anyone else) have to make the correct choice of profession, and that this choice must be retained for the rest of their lives (Wrightsman, 1994), indecision or change of profession being regarded almost as mental disorder requiring treatment (Krumboltz, 1992). This search for the correct profession illustrates a concept widespread in society that identity (professional, in this case) is something that is relatively ready within individuals, static and immutable, therefore requiring only to be "discovered" for the best choice of career to be selected. . . .
The meaning of change, therefore, is revealed in the plan through which the subjects decide to take account of their lives. From this standpoint, we can propose a new interpretation of what those interviewed named maturity: it is the capacity to make choices (their own) and to make plans, professional plans. When speaking of their own or authentic choices, this is not to say that unconscious factors or social or economic pressures may not exist which override the choice in some form. It is intended to emphasize, however, the active nature that the subjects can have in constructing their history, even though they may not have full control or knowledge of their motives.

The rich and exhaustive descriptions about the seven Brazilians' experiences with voluntary career change in Teixeira and Gomes (2000) extended previous career theories and provided new knowledge about a phenomenon that was understudied. The authors' detailed illustrations of the method, analysis results, and their reasoning help the reader understand how their interpretations were derived from the data. More exemplars of phenomenological studies can be found in Wertz (2005), McLeod (2001), and the Journal of Phenomenological Psychology.

CONSENSUAL QUALITATIVE RESEARCH
In this section, we will briefly introduce a relatively newly developed strategy of inquiry, consensual qualitative research (CQR), followed by a discussion of exemplar studies. CQR is a systematic way of examining the representativeness of results across cases through the process of reaching consensus among multiple researchers.
Hill, Thompson, and Williams (1997) developed CQR based on the grounded theory (Glaser & Strauss, 1967; Strauss & Corbin, 1990), comprehensive process analysis (Elliott, 1989), and phenomenological approaches (Giorgi, 1970, 1985, as cited in Hill et al., 1997), as well as feminist theories (e.g., Fine, 1992; Harding, 1991, as cited in Hill et al., 1997). There are many common premises and differences between the CQR and these other qualitative approaches (see Hill et al.; Hoshmand, 1997 for a review). Yet, as Hoshmand noted, the initial guidelines for CQR (Hill et al., 1997) "did not articulate the implicit philosophical perspective on which the communal processes of CQR are based" (p. 601). For these reasons, CQR has been described as a "generic" qualitative approach (McLeod, 2001, p. 147). Recently, Hill and her colleagues (2005) and Ponterotto (20056) attempted to locate the philosophical stance of this method in order to link the gaps among the ontology, epistemology, and methodology for CQR.
The primary components of CQR include gathering data through interviews and open-ended questions, describing the phenomenon of interest using words (not numbers), using a criterion-based sampling method, and understanding the parts of the experiences in the context of the whole. Usually a small number of cases (8 to 15) are studied. A primary team of three to five researchers independently and inductively analyze narratives of participants' experiences and then form consensus on ways to interpret the data. Through this process, multiple perspectives are included in the data analytic decisions. Also, domains and core ideas (abstracts) are identified from the data. Categories are developed by cross-analyzing the consistencies in the core ideas within domains; the terms—general, typical, variant, and rare—are used to characterize the occurrence rate of the categories. One or two auditors examine the consensual judgments on the domains, core ideas, and categories in order to ensure that important data were not overlooked (Hill et al., 1997; Hill et al., 2005).
The following key elements of CQR will be illustrated in greater details: (1) researchers, researcher biases, and training of the research team; (2) participant selection; (3) interview (data collection) and transcription; and (4) data analytic procedure.




RESEARCHERS, RESEARCHER BIASES, AND TRAINING OF THE RESEARCH TEAM
Hill and colleagues (1997) stress the importance of clearly describing the researchers because the data analyses of the CQR rely on the consensual process of the research team members. Although the composition and the dynamics of the team are essential, there is no absolute correct type of team composition (e.g., faculty members, senior practitioners, graduate or undergraduate students). Researchers need to clearly consider the various possibilities of the team composition and be very careful in selecting the team members, considering possible power differences among the group members, types of expertise, and the level of commitment needed from each team member. Before the data collection and analytic process begin, researchers receive appropriate training in conducting interviews and data analyses. It is also crucial that the researchers bracket (i.e., set aside) their biases and the information gained from the literature so that data Can be approached from a fresh perspective and allowed to "speak" for themselves.

PARTICIPANT SELECTION
CQR adopts the criterion-based sampling method (Goetz & LeCompte, 1984, as cited in Hill et al., 1997), for which the criteria of selecting the participants should be clearly stated. This method ensures that the participants have had some depth of experiences with the phenomenon of interest and can provide meaningful information for the purpose of the study. Usually, 8 to 15 cases are studied intensively when using CQR because: (1) this sample size, albeit small, usually provides a sufficient number of cases for the researchers to examine variability and consistencies across cases; and (2) additional cases typically add minimal new information (Hill et al., 1997). However, the sample size should also be determined by the homogeneity of the sample and the amount of data collected (e.g., length of the interviews) from each participant (Hill et al., 2005).



INTERVIEW (DATA COLLECTION) AND TRANSCRIPTION
An interview protocol can be developed based on literature review, conversations with the population of interest, and researchers' personal reflections (Hill et al., 2005). Pilot interviews are conducted to evaluate the adequacy of the interview protocol (at least two pilot interviews are recommended by Hill and colleagues). The actual interviews could be conducted over the phone or face to face. Previous CQR researchers also have used survey (open-ended questions) or email format to collect data (see Hill et al., 2005).
Immediately following the interviews, the interviewer(s) should record memos (e.g., impression of the interviewee, comments about the flow of the session) that later may be used to facilitate data analysis. Once the interview is completed, it should be transcribed verbatim with identifying information omitted and unnecessary nonlanguage utterances ("urn," "ah") and fillers ("you know") deleted. Copies of the transcripts could also be reviewed by the interviewees for additions, corrections, or clarifications. This step is called "member check," and it enhances the credibility and trustworthiness of the data (Lincoln & Guba, 1985).

DATA ANALYTIC PROCEDURE
Hill and colleagues (1997) delineated the step-by-step data analytic procedure and divided it into the following sequential stages: (1) identify domains, (2) summarize core ideas, (3) construct categories from the cross-analysis, (4) audit the analysis results, (5) check the stability of the results, and (6) chart the findings. Specifically, these procedures were summarized in the update article about the CQR method as follows:
Domains (i.e., topics used to group or cluster data) are used to segment interview data. Core ideas (i.e., summaries of the data that capture the essence of what was said in fewer words and with greater clarity) are used to abstract the interview data within domains. Finally, a cross-analysis is used to construct common themes across participants (i.e., developing categories that describe the common themes reflected in the core ideas within domains across cases). . . . The auditor . . . provides detailed feedback at each stage of the analysis process (e.g., creating do¬mains, constructing core ideas, creating the cross-analysis). . . . For a stability check . . . after the domains and core ideas were completed for all of the cases, at least two cases be withheld from the initial cross-analysis and then used as a check to determine whether all of the data for these cases fit into the existing categories and whether the designations of the general, typical, and variant changed substantially with the addition of the two new cases. . . . Hill et al. (1997) recommended charting the results to depict visually the relationships among categories across domains, particularly for data representing sequences of events. (Hill et al., 2005, pp. 200-202)